Download Landscapes and Their Ecological Components

213-A1678 8/6/01 10:07 AM Page 435
SECTION I / THE BIOSPHERE
ALMO FARINA
Lunigiana Museum of Natural History
Aulla, Massa, Italy
Landscapes and
Their Ecological
Components
Landscape ecology is a discipline that studies
environmental complexity, concentrating mainly on
an analysis of the importance of spatial relationships
between the various components (individuals,
populations, communities, and land mosaics) of the
real world. Defined as the spatial representation of
an ecosystem, a landscape is composed of many
superimposed environments as perceived by various
species or modified by ecological processes such as
flooding and fires. The way in which an organism
perceives the environment that it inhabits is species
specific, which allows us to talk of a geobotanical
domain, an animal one, and an anthropic one. These
complex landscapes are measured spatially by
geographic information systems, and the resulting
data are analyzed using a combination of geostatistics
and fractal mathematics. Landscape ecology is
used mainly for the management of human-scale
landscapes and, more specifically, for the analysis,
management, and conservation of nature; these offer
the most fruitful application of the knowledge that
landscape ecology provides about the workings of
land mosaics.
◗
................................................
Introduction
During the past few decades, it has become increasingly
clear that in order to acquire a more thorough understanding of the complex processes that take place in the biosphere and, more generally, in the ecosphere, the spatial
dimension of ecosystems must be broadened. An appreciation of the extent of the complexity of life on Earth has become widespread among both scientific researchers and
the general public, and the globalization of environmental
problems has made it necessary to broaden the scale of the
analysis and of the interpretation of the phenomena involved. The broadening of the spatial scale has also led to
the reassessment of the temporal one and, as a result, to the
creation of a powerful and flexible analytical tool in which
the spatial and temporal scales can be calibrated according
PART TWO / DISCOVERY AND SPOLIATION OF THE BIOSPHERE
to the system being considered. Ecological research, which
until the beginning of the 1980s was rooted in concepts of
environmental homogeneity, of equilibrium within the ecosystem, and of the separation of man from environmental
processes, has undergone a profound change during the past
decade. Indeed, when the spatial scale was broadened, ecological research began to view the environment as a heterogeneous system in which the variety of ecological processes
are expressed as a state of dynamic equilibrium (homeoretic flows) and in which the processes governed by man
become an intrinsic part of environmental mechanisms.
Landscape ecology studies environmental complexity, concentrating mainly on an analysis of the importance of spatial relationships between the various components of the
real world.
Heterogeneity appears to be the predominant pattern in
435
213-A1678 8/6/01 10:07 AM Page 436
ALMO FARINA
most landscapes, and increasing importance is attached to
the role that it plays in determining ecological processes. A
landscape contains heterogeneous characteristics that are
expressed in discrete entities known as patches, which in
turn make up a mosaic that is structurally and geographically distinct. The salient characteristics of these patches,
such as dimensions, shape, type of vegetation, biological
richness, abundance of organisms, and flow of nutrients,
contribute to the organization and maintenance of the complexity of landscapes.
◗
................................................
Landscapes
A landscape is defined as the spatial representation of an
ecosystem. Thus, rather than an entity being described entirely in terms of its function (an ecosystem), it is defined in
terms of its spatial characteristics (a landscape). The concept of landscapes represents a relatively new approach
and one that is also rich with significance. Indeed, as the
geographic dimensions are expanded, ever more processes,
together with the increased biodiversity they bring, are included. A landscape, in fact, is a system of environments,
each one of which is species specific and delimited by the
species. Each species recognizes the boundaries of the environment it inhabits, identifies those characteristics that favor its existence, and reacts to the environment on a speciesspecific scale. A landscape is therefore the sum of many
species-specific systems that can be superimposed to a varying degree and that contribute to the complexity of the environment. This vision of the environment is characterized
by a large quantity of information that results in a set of
retroactive mechanisms that help to regulate and calibrate
the various processes and that make it possible for many
organisms to survive and to perform multiple roles. All this
tends to promote an autopoietic capacity, or the “creative”
ability of the system to adapt to new conditions through
homeoretic and homeostatic mechanisms.
When studying a landscape, it is important to be able to
consider the components that are of immediate interest
without losing sight of the whole. Such an approach is necessarily transdisciplinary, and analyzing a landscape does
much to increase our understanding of the ecological complexity of a geographic area. A landscape can vary in size
from a few centimeters to tens of kilometers. In fact, the
term immediately brings to mind a large area, and in almost
all Western cultures it is used to denote a scenographic or
geographic reality. Nevertheless, every species perceives its
own particular landscape in its own species-specific way,
which can even change according to the physiological context in which that organism finds itself at any moment. The
landscape is therefore perceived by man as his own environment. Consequently, the study of the way in which it functions is called landscape ecology; indeed, this dimension is
436
the nearest to the complex processes that, through his own
culture, man has applied to a natural substratum.
As a result of the variety of cultural and scientific approaches, there are many definitions of a landscape as perceived by man. It is the general character of a region (Humboldt, 1807); it is a heterogeneous area composed of a
cluster of interacting ecosystems that reoccur in a similar
manner throughout a region (Forman and Godron, 1986);
it is the sum of physical, ecological, and geographic entities
that embrace and bring together all natural and human processes and their patterns (Naveh, 1987); it is a particular
configuration of topography, plant cover, use of the soil, and
human settlements that are delimited by natural and cultural processes as well as by anthropic activities (Green
et al., 1996); and W. Haber (1996, personal communication)
defines a landscape as an area of land that we perceive without paying particular attention to any one component that
seems familiar to us.
Currently, most landscape ecology is concerned with
landscapes that have been modified by man, even though its
analytical principles and instruments can be validly applied
to the study of landscapes that have not been subjected to
the direct influence of man. Landscape ecology therefore
studies complex systems and indeed often requires that special reference be made to a particular organism or group of
organisms (Turner et al., 1995). Consequently, a landscape
perceived by human beings is quite different from a landscape perceived by coleoptera (Wiens and Milne, 1989) so
that when one talks of a human landscape one must necessarily consider the cultural component that is associated
with it.
The Contribution of Other Disciplines
to Landscape Ecology
Precisely because landscape ecology is a complex discipline, its theoretical corpus is the result of bringing together and integrating a series of approaches and applications, among which the theory of island biogeography
(MacArthur and Wilson, 1967) and geographical ecology
(MacArthur, 1984) occupy preeminent positions. These
theories were the first to emphasize the extent to which the
degree of isolation and the dimensions of an island determine species richness and abundance. Islands that are
small and distant from the mainland have a smaller number
of species compared to those that are larger and nearer to
the mainland. Furthermore, there is an inverse correlation
between the risk of extinction and the probability of colonization. The smaller the island, the greater the risk of extinction, whereas the probability that another species will
colonize it is correspondingly less. This theory can also be
usefully applied to so-called islands on land, environments
that due to the process of fragmentation have been reduced
in size and have been surrounded by other environments
that are hostile for given species.
VOLUME IV / THE LIVING WORLD
213-A1678 8/6/01 10:07 AM Page 437
LANDSCAPES AND THEIR ECOLOGICAL COMPONENTS
Landscapes and Heterogeneity
A landscape consists of two fundamental components:
the geographic or chorological component and the topological or functional one. In fact, these two components cannot be separated without drastically reducing the amount
of information that is obtained when they are considered together. By definition, landscapes are heterogeneous environments: That is, they consist of various elements that are
interrelated in different ways. This heterogeneity is a very
important pattern since it in fact determines the ecological
diversity of any region. The heterogeneity might be expressed as physically identifiable structures, such as woods
and land under cultivation, or as processes such as variations in the thermal characteristics of slopes or the degree
of the environment’s species-specific suitability. At any rate,
the degree of heterogeneity varies according to the spatial
arrangement of the single component parts. Space is therefore an important component in determining the extent of
the diversity of the various forms of life. It follows that the
spatial arrangement of the various entities and their processes and relevant patterns can be considered as one of the
main paradigms for introducing the spatial dimension as an
element that is important per se. The presentation of landscapes as components in a “nest” hierarchy (one in which
the highest level contains all the other levels) that incorporates the concept of scale has been a determining factor
in linking the various paradigms (Allen and Starr, 1982;
O’Neill et al., 1986) and the various theories (Delcourt and
Delcourt, 1988).
At the same time, a more satisfactory approach to dealing with complexity has been provided by non-Euclidean,
fractal geometry (Mandelbrot, 1975), which is able to measure the complexity and the relative patterns of many natural systems and to establish relationships between elements
that are only distantly connected. New theories of heterogeneity (Kolasa and Pickett, 1991) and the role of disturbance regimes in ecological processes (Pickett and White,
1985) represent significant steps forward and thoroughly
incorporate the paradigm of ecotones (Hansen and Di Castri, 1992) and the processes associated with them, such as
connectivity and connection (Merriam, 1984) and metapopulation models (Gilpin and Hanski, 1991), into landscape ecology. The recognition of the existence of a heterogeneous structure in landscapes has also made it possible to
extend and adapt Pulliam’s (1988) source and sink demographic model and to redefine the roles played by single
patches in a landscape.
The Study of Landscape Ecology
The complexity of the issues they deal with, together with
the fact that they make reference to a range of disciplines,
has led to landscapes being viewed as entities that change
character according to the approach that is adopted. The
main texts on landscape ecology (Forman and Godron,
PART TWO / DISCOVERY AND SPOLIATION OF THE BIOSPHERE
1986; Naveh and Lieberman, 1994; Forman, 1995; Zonneveld, 1995) have drawn ample attention to this characteristic while nevertheless expressing very different and often
divergent visions of the discipline. Indeed, whereas Naveh
and Lieberman have an anthropocentric, globalizing vision,
that of Zonneveld is more geographical and concerned
with typological classification and the approach of Forman
and Godron is more geometric and views space on a large
scale as the central factor that conditions processes.
It is time to reconcile and integrate the various concepts
of what a landscape is, bearing in mind the different biological components (Farina, 1998). Historically, this discipline derived from the human geographical sciences, and
the first description of families of patterns and processes
was strictly linked to the anthropic sphere. Recent studies,
mainly in North America, on the spatial arrangements of
patterns and on processes that concern the soil, vegetation,
and the animal component apart from the anthropic one,
have resulted in a huge leap forward. There are indeed three
perspectives to be considered in landscape ecology: the human one, the geobotanical one, and the animal one.
The human perspective allows us to dismantle and reassemble a landscape according to the functional entities that
are relevant to human beings. The geobotanical perspective
considers the spatial distribution of biotic and abiotic components, from the landscape of the soil to that “perceived”
by plants and to the distribution of vegetal entities such as
woods and grassland. Where plants are concerned, “perception” is taken to mean the ability to absorb information from
the surrounding environment. This is directly related to the
extent of adaptation, to the rate of colonization and extinction, and to the resistance to natural and man-made stresses.
The animal perspective considers the way in which each
species perceives its own environment. It is obvious that a
butterfly cannot use the same environmental parameters as
a deer or a fish. Its life will therefore take place in and impact a precise part of a landscape. The combination of the
interactions expressed by an animal community constitutes
an important conditioning factor of the landscape seen as a
complex whole. Man, performing the double role of biological species and intelligent organism, should also be considered in this context.
These perspectives are not in conflict because each one
of them is concerned with exploring a domain of patterns
and processes that are in fact components of the entire biological and ecological system. Furthermore, there are many
more points in common, such as the perception of space
and the spatial arrangement of patterns and processes, than
there are differences. Thus, the common strategy adopted
by landscape ecologists is to insert these patterns and processes into the most suitable spatial and temporal scale.
The human dimension of a landscape is probably the
most complicated in that it superimposes man’s animal and
cultural components. This dimension relates to processes
437
213-A1678 8/6/01 10:07 AM Page 438
ALMO FARINA
that have a very broad spatial and temporal scale. The biological dimension of man can be compared to that of animals, but the cultural component of humanity is unique,
and it is this component that has the greatest interactive
impact on a landscape, especially since man is inclined to
use technologies that overcome the physical and biological
limits of the natural systems in which he operates.
The three perspectives outlined previously cannot be
considered separately, but their sectional nature is linked to
the ability to adopt an approach that is interdisciplinary or
even transdisciplinary. A landscape is so complex that even
this approach does not provide explanations for all the processes; but it can shed new light on this complexity and
define the limits of its own ability to provide explanations.
Landscape ecology is therefore one of the most promising ecological disciplines, and although it is highly differentiated and wide ranging, it is capable of expressing a single
“spirit” linked to the finite dimension of the area of study.
This discipline allows a more precise superimposition of
information from the real world onto the virtual processes
that today’s postindustrial culture is promoting through
globalization.
Currently, there is a high risk of landscape ecology being
considered entirely from the anthropocentric point of view,
and this might result in its being seen as a dogmatic discipline without either theoretical foundations or experimental backing. Furthermore, it seems unacceptably reductive
to consider landscape ecology simply as a form of broadscale ecology. In fact, one of the major strengths of this discipline is its ability to transfer information through various
families of processes that take place on different spatial and
temporal scales and to link knowledge from precise ecological research to environmental and landscape planning
activities.
◗
................................................
Theories and Models Embraced
by Landscape Ecology
The hierarchical theory and the percolation theory combine with two models of population dynamics—the metapopulation model and the source and sink model—to create a powerful referential context for landscape ecology.
Although they were formulated in various contexts, these
theoretical elements share a common interpretation of the
complexity of landscapes. They have played an undeniably
important role in creating a homogeneous disciplinary corpus that offers a paradigmatic bridge over the complexities
of landscapes.
The Hierarchical Theory and
the Structure of Landscapes
The hierarchical theory (Allen and Starr, 1982; O’Neill
et al., 1986; Allen and Hoekstra, 1992) is a useful instrument
438
for exploring numerous patterns and processes through
various scales in space and time. Considering complexity as
an attribute that is intrinsic to a landscape, the hierarchy
paradigm explains how the various components located on
certain scales enter into contact with other ones that are
visible on different scales of resolution. The hierarchical
theory views a system as a component in a larger system
that consists of subsystems. As one moves from one system
to another, the characteristics of the phenomena change—
for example, the classification of a landscape as one goes
from lower to increasingly higher levels in the hierarchy:
ecotope (the basic unit in a landscape consisting of biotic
and abiotic elements); microchore (the spatial distribution
of ecotopes); mesochore (the environmental system composed of a group of microchores); macrochore (a mosaic of
landscapes); and megachore (a group of geographical elements covering several kilometers).
A watershed is an example of a hierarchical system composed of subbasins that in turn are composed of even smaller
basins. A system exists independently of its components
and is generally able to organize itself and to transmit information; in other words, it is able to exist as a cybernetic
system. A landscape exhibits its own type of complexity, and
in order to understand it fully it is necessary to focus on a
certain organizational level. There are innumerable hierarchical levels and thus an equal number of systems that are
nested inside them in one way or another. The behavior of
a given subsystem conditions nearby systems both above
and below it. The speed with which the processes unfold
and thus the scale in time are specific to each level. When
going from one level to another, it is therefore necessary to
adjust the resolution. The hierarchical theory contributes
much, for example, to an understanding of the effects of
various disturbance regimes. Thus, fire is a highly destructive element for the “tree system,” but it is an element that,
ecologically speaking, promotes “creativity” for the forest
as a whole in that it establishes new conditions for many
species of plants and animals.
The Percolation Theory
The percolation theory was formulated as a result of
studies of the behavior of fluids in a medium (Stauffer, 1985)
and has recently been used, with interesting results, in the
creation of landscape models (Gardner et al., 1987). Unlike
diffusion processes in which every particle in a liquid moves
in all directions, the percolation process takes place in finite regions of the medium, which the fluid does not leave.
When the fluid occupies about 0.5928% of the entire surface of the matrix, it has reached what is known as the percolation threshold (Ziff, 1986); such a state guarantees, for
example, that the cells in a matrix will be in contact with
each other from one side of the matrix to the other. This
theoretical limit is nevertheless an important concept with
regard to studying real patterns, such as the expansion of
VOLUME IV / THE LIVING WORLD
213-A1678 8/6/01 10:07 AM Page 439
LANDSCAPES AND THEIR ECOLOGICAL COMPONENTS
an urban area or the spread of a wood as a result of secondary succession.
The relevance of this theory to the study of the behavior
of landscapes is clear if one considers that around the critical threshold, contagion, forest fires, and demographic explosions of pathogenic agents exhibit their initial states of
diffusion (Turner, 1987). Furthermore, this theory can be
usefully applied to the study of animal behavior in a heterogeneous system. Indeed, the percolation process has been
thoroughly tested with neutral models (i.e., square matrices with randomly filled elements). Neutral models assume
a state of complete spatial independence between the various cells, and the habitats do not exhibit any autocorrelation. These models consider both the scale of resolution
and the behavior of the single cells of the matrix and, for
example, simulate the perception that the various species
have of the environment that surrounds them. Since the perception of an environment is species specific, it is obvious
that each species will perceive it as a percolating habitat
(i.e., one that is more or less connected). These models can
be used to predict when an environment that is in a state of
fragmentation will begin to lose important characteristics,
such as connectivity.
The Metapopulation Model
Increasing forest fragmentation is a general trend in natural environments that has created fragments of isolated
woodland. For the remaining populations, fragmentation
has increased the risk of extinction. It can be assumed that
when a population lives in a fragmented and heterogeneous
environment, it perceives the landscape in which it lives as
a mosaic in which hospitable environments alternate with
ones that are more or less hostile, and that contact between
the various subpopulations is ensured only through processes of emigration or immigration. The risk of both local
extinction and recolonization depends on the extent and
existence of an exchange of individuals. The success of recolonization depends on many factors, including the ability
to disperse, and these populations are therefore considered
to be components of a metapopulation. The term metapopulation was first used by Levins (1970) to describe a
population of populations (Gilpin and Hanski, 1991; Hanski and Gilpin, 1991). Metapopulations are systems in which
the rates of extinction and colonization create flows of individuals that ensure a genetic connection between the
subpopulations.
The metapopulation model is closely related to the theory of island biogeography (MacArthur and Wilson, 1967),
but it is distinct from it in that the terrestrial systems in
which the metapopulations develop are less hostile. In the
theory of island biogeography, the distance between the
various islands is the main barrier to dispersal, and the distance from the mainland is one of the factors that determine the rates of extinction and recolonization. In the meta-
PART TWO / DISCOVERY AND SPOLIATION OF THE BIOSPHERE
population model, however, there is a constant relationship
between the various subpopulations; as a result, the exchange of individuals and therefore of genetic inheritance
is maintained at a high level. The metapopulation model can
be usefully applied to fragmented areas in order to analyze
their dynamics and predict the probability of extinction or
recolonization of isolated populations.
The Source and Sink Model
The source and sink model was created by Pulliam (1988)
in order to study the responses to environmental conditions
among various species of birds. According to this model,
which can be adapted very effectively to heterogeneous environments in which the resources are unevenly distributed,
a source population is one in which the ratio between births
and deaths is always above unity. In other words, in a source
population there is always a surplus of individuals that will
tend to leave it. On the contrary, in a sink population the
ratio between births and deaths is below unity, and such a
population can survive only if new arrivals from outside
add to it. This model can also be used to classify patches in
a heterogeneous environment according to their ability to
sustain a source population or a sink population (Fig. 1).
As a general rule, the larger patches in an environment will
tend to act as source ones to the smaller patches. Whether
a patch is a source one or a sink one will also depend on
source
sink
FIGURE 1 The source and sink model is used to analyze the
distribution of the populations of a species in a heterogeneous
environment not according to their density but to the ratio between births and deaths. A source area is defined as one in
which there are more births than deaths, and a sink area is one
in which the populations cannot survive because the number
of births is less than the number of deaths. The continuing
presence of a population in a sink area is guaranteed only by
the arrival of individuals from source areas.
439
213-A1678 8/6/01 10:07 AM Page 440
ALMO FARINA
the seasonality of the resources and on the role that these
resources play in the survival of a species at the various
levels of aggregation. Indeed, this model can be usefully
applied in fields outside population demography. Thus, if
the resources in a given patch are such as to guarantee the
survival of a species, that patch can be defined as a source
patch.
There are variants of the source and sink model. For example, Watkinson and Sutherland (1995) coined the term
pseudo-sink to describe an environment that appears to exhibit sink characteristics but in which, even if immigration
comes to a halt, the population will stabilize at a minimum
level, without disappearing. Furthermore, there are special
cases in which environments with sink characteristics are so
inviting to various species that they take on the appearance
of source environments (Pulliam, 1996). This is the case with
many environments that have been modified by man and
that offer abundant resources but that do not, for example,
guarantee adequate protection against predators or where,
in the sensitive period for a given species, man produces a
disturbance that interferes with successful reproduction or
even makes it impossible. It must be pointed out that this
model does not always allow one to deduce the qualities of
an environment, especially if the demographic criterion is
applied. There are times when a sink environment can be
inhabited by more individuals than a source one, and there
are populations in which most of the individuals inhabit
sink patches.
◗
................................................
Ecotones
The heterogeneous nature of landscapes ensures that there
is contact between patches with varying characteristics
(Pickett and White, 1985; Hansen et al., 1992). These junction zones are called ecotones. Clements (1905) described
ecotones as zones of tension in which the species of neighboring communities meet. Clements was the first to use
the term, which is based on the two Greek words, oi\ko~ and
tovno~.
Ecotones can be defined in many different ways, according to whether one is considering patterns, processes, or
fully fledged habitats. They are therefore also habitats that
serve as meeting points for species that are able to fully exploit the frontier zone characteristics of these environments (Ricklefs, 1973). In fact, even though the concept of
ecotones is simple, their scalar characteristics complicate
the task of studying them. For example, the closer one gets
to an ecotone, the more difficult it becomes to distinguish
what was clearly visible from a distance. Furthermore, the
study of ecotones is complicated not only by these scalar
characteristics but also by the fact that every biological species perceives and interacts with the environment in a spe-
440
a
b
c
a
b
c
d
FIGURE 2 An ecotone (in red) is a transitional zone between different environments and is perceived in a speciesspecific way. Thus, there are species that do not perceive an
ecotone (a), species that live entirely within it (b), species that
see it as a hostile area and look on it as a barrier (c), and species for which the ecotone is part of their habitat (d).
cies-specific way, with the result that the position of ecotones varies from species to species (Fig. 2). Thus, ecotones
that can be identified by analyzing an aerial photograph or
a geological map do not always exist as far as every species
is concerned.
Ecotones are important structures with regard to the
functioning of landscapes. Indeed, an exchange of nutrients,
water, spores, seeds, and animals takes place within these
juncture zones between different patches. The fact that they
are transitional zones also makes them excellent indicators of climatic changes linked to the increase in CO2 and
atmospheric pollution. Ecotones also have high rates of
primary and secondary productivity since they generally
benefit from the various characteristics of their constituent
environments. In environments that have been modified by
man, ecotones that are composed of copses, hedgerows,
rows of trees, and the edges of fields are often refuge areas
for many species that are sensitive to the disturbance regime created by man.
Along rivers and in lagoon areas, ecotones are junction
zones between terrestrial and aquatic environments, intercepting the flow of nutrients moving from one to the other.
The importance of ecotones has been recognized for a long
time (Clements, 1897; Livingston, 1903; Griggs, 1914). Although authoritative ecologists have emphasized their importance at various times (Weaver and Clements, 1929;
Odum, 1959; Daubenmire, 1968; Ricklefs, 1973), only recently have systematic studies on the subject been carried
out (Di Castri et al., 1988; Naiman and Decamps, 1990; Holland et al., 1991; Hansen and Di Castri, 1992; Risser, 1995).
It now seems that the study of landscapes with reference
to ecotones is capable of shedding much light on the way in
which matter, energy, and individual organisms interact with
a heterogeneous environment and thus with the very complexity of landscapes. It is no coincidence that the edges of
mosaics have a great influence on the richness and the move-
VOLUME IV / THE LIVING WORLD
213-A1678 8/6/01 10:07 AM Page 441
LANDSCAPES AND THEIR ECOLOGICAL COMPONENTS
ment of organisms, on the flow and accumulation of matter
and energy, and on the spread of disturbances. In order to
understand ecological processes, it is therefore essential to
study ecotones, especially when the object of study is landscapes, which by definition are a part of the real world
that we have an interest in investigating. Ecosystem ecology, which focuses mainly on the mechanisms that operate
within an ecosystem and which analyzes the internal parts
of homogeneous areas, differs considerably from landscape
ecology, which is concerned with the functioning of ecological systems operating within a geographical context.
It is possible to classify ecotones according to their origins. Holland (1988) suggested four types: ecotones that
are created and maintained through human disturbance
(hedgerows, plow land, and hay fields); those that are created and maintained by natural processes, such as ecological succession or the bridge between terrestrial and aquatic
environments that results from beavers’ dams; those that
are created by natural processes and maintained by man,
such as coastal lagoons or riparian woodland; and those
created by man and maintained by natural processes, such
as the marshy areas around artificial water reserves.
Generally, there are substantial differences between natural and man-made ecotones. In the former, the transition
from one patch to another is gradual, whereas in the latter
it is sudden and clearly defined.
Characteristics of Ecotones
Although not easy, it is nevertheless possible to identify the structure and functions of an ecotone. An ecotone
covers the whole junction zone between one system and
another. Its shape can be linear, as in man-made ecotones,
circular, or convoluted, and it will determine the rate of
transfer of environmental information, energy, and matter
through the ecotone.
The biological structure is determined by the biomass or
density of the dominant organisms. The structural contrast
is determined by the difference between the structure of
the ecotone and that of the neighboring environments. The
internal heterogeneity is determined by the variability of
the changes within the ecotone. The density of the ecotone
is determined by the ecotonal surface area of a given area.
There are also other important functional variables such as
stability, which is a measure of the degree of the ecotone’s
susceptibility to change. Resilience, on the other hand, is a
measure of the ecotone’s ability to return to its previous
state following a disturbance.
Both internal and external factors condition the organization of ecotones. The external factors are determined by
processes that act on landscapes, and the internal factors
represent elements of discontinuity. Many animal species,
for example, create conditions that are hostile for others;
furthermore, on an intraspecific level, territoriality can dis-
PART TWO / DISCOVERY AND SPOLIATION OF THE BIOSPHERE
play ecotonal characteristics that are similar to a change in
the vegetational conditions within a mosaic. Some species
of plants modify the pH balance of the soil to such an extent that they create ecotonal conditions that are more or
less hostile for other plants. The shade produced by a canopy can reduce the ability of other plants to grow. Many invasive species, such as the Jerusalem artichoke (Helianthus
tuberosus), grow so densely that they prevent other species
from establishing themselves.
Many animal species create ecotones as a result of their
activities. An example is a small mountain field mouse, Microtus nivalis, that lives in grasslands above the treeline. Its
incessant search for food, which does not even stop in winter, has a major impact on the distribution of the bilberry
(various species of Vaccinium), whose evergreen stems it
eats, especially in the cold season (Farina et al., 1986). Similarly, as a result of digging holes and making molehills and
nests above ground level, moles, ants, and termites modify
the plant cover and even change the microtopography of
the soil.
In summary, ecotones can be found both on a megascale, such as a bioclimatic region, and on a microscale, such
as an area covering just a few square centimeters of ground.
The role that they play in a landscape is fairly clear. They
represent semipermeable membranes that allow processes
of passive diffusion, such as the movement of air or water
masses carrying matter or organisms, and also processes of
active diffusion, by which organisms move biomass from one
part of a landscape to another, thus modifying the environmental information as well.
There is no doubt that the presence of ecotones contributes to the stability of a system by guaranteeing a complexity that reduces the risk of sudden deterioration and
the loss of the system’s autopoietic capacity. Thus, a riparian ecotone (consisting of the vegetation along the bank of
a river) reduces the effects of flooding by retaining, among
other things, large quantities of organic substances that are
then used in the food chain. Many organisms, including most
amphibians, live only in ecotonal zones. The considerable
species richness of the coral reefs, for example, is a function
of the ecotonal characteristics of these structures. Furthermore, the greatest concentrations of plankton species are to
be found in areas in which large masses of seawater meet.
Animal richness and abundance are greater along ecotones than in neighboring areas, but the linear nature of
these environments also increases the risk of predation. In
other words, these structures can play a variety of roles,
even if by themselves they do not represent paradigms that
are fully able to explain the complexity of landscapes or,
therefore, the way in which they function. However, if this
paradigm is coupled, for example, with the source and sink
model or with the hierarchical theory, the combination is
capable of yielding a great deal of information.
441
213-A1678 8/6/01 10:07 AM Page 442
ALMO FARINA
◗
................................................
Cultural Landscapes
Most terrestrial landscapes have been profoundly modified
by man, who has built towns, cut down forests, changed the
vegetation, constructed roads and motorways, and brought
about changes in the climate. In terms of the perspective
adopted by landscape ecology, these modifications affect
the spatial arrangement of patches, modify the fragmentation of large forest patches, and increase the number of transitional zones or ecotones.
Such modified systems exhibit an increase in both the
fragmentation of the original forests and environmental
heterogeneity. In these cases, landscape ecology is a discipline that offers an increasingly useful analytical tool in that
it allows ecologists to study patterns and processes on different scales, including environments that have been modified by man (Fig. 3). Therefore, this approach has an increasingly important role to play in many forms of human
activity that are directly related to the environment and its
characteristics. Indeed, landscape ecology is able to study
ecological processes over a broad range of spatial and temporal scales that reflect the complexity of the systems. By
applying this discipline, it is possible to manage key species,
residual forest zones, and edge networks and to influence
or decide on urban development policies. Specifically, it can
be applied to the problems of many landscapes that, having
been modified by man, have acquired a high value in terms
of complexity, biodiversity, and scenic beauty. These systems are called cultural landscapes; that is, landscapes that
have been subjected to a man-made disturbance regime for
a long time and in which the character and spatial arrangements of the patches are the result of complex retroactions
between man and nature.
Cultural landscapes reflect the interactions between man
and the natural environment and are complex phenomena
with identities that are both tangible and intangible (Plackter and Rossler, 1995). In 1991, the secretary of the United
Nations Educational, Scientific, and Cultural Organization
set out a series of guidelines for identifying these landscapes. They generally exhibit a complex structure consisting of a fine-grained mosaic in which the physiotopes, or
physical units of the landscape, are markedly localized and
are used for specific and appropriate purposes (e.g., agriculture, forestry, and grazing). In upland and mountain areas, the slopes have often been terraced in order to facilitate the working of the land and, at the same time, to
reduce erosion and retain nutrients (Fig. 4).
Cultural landscapes might represent a useful model for
testing ways of ensuring that man’s presence does not destroy resources or cause irreversible disruption to the environment. Currently, such an aim might seem utopian,
granted the universal presence of technology, but the lesson that cultural landscapes teach us cannot be ignored and
FIGURE 3 A cultural landscape in Tuscany. Created and structured as a result of close interactions between
agriculture, cultural heritage, and environmental dynamics, it requires a constant input of energy (in the form of
work) from man.
442
VOLUME IV / THE LIVING WORLD
213-A1678 8/6/01 10:07 AM Page 443
LANDSCAPES AND THEIR ECOLOGICAL COMPONENTS
tends to result in unusually frequent fires, which reduce ecological complexity. Indeed, although most plants display a
high tolerance to fire, it is precisely the greater frequency of
this disturbance that prevents ecological succession from
occurring and consequently increases the risk of erosion
during periods of high rainfall.
◗
................................................
Ecosystem Management
this model should be considered for many regions of the
earth. Every cultural landscape is the product of the interaction between man and the environment over a very long
period; thus, when human activity decreases, as happens increasingly often with the abandonment of farms in mountain lands, the fragility of this system becomes evident and
cultural landscapes lose a part of their complexity. For example, once terraced land is abandoned, the terracing deteriorates rapidly.
Most cultural landscapes are different from each other,
but they generally exhibit more linear structures and often
have more open spaces compared to undisturbed landscapes. However, in some cases the opposite happens, such
as in desert oases, in which the density of palm trees is
greater than that in the surrounding desert. Cultural landscapes are generally the result of interaction between the
resident population and the natural environment, but even
nomadic populations such as the Mongols must have played
a major role in shaping the character of the steppes. Entire
regions, such as the Mediterranean, can be considered to be
cultural landscapes, and the modifications brought about
by man can have irreversible effects both on biodiversity
and on ecological diversity. Therefore, many people are
worried that these important phenomena, which also attract much tourism, will disappear. When evaluating what
often appears to be natural and spontaneous, it is important to proceed with care; in regions such as the Mediterranean, biodiversity has been reduced over thousands of
years, and there are no more new species to fill the ecological niches left empty by species that failed to adapt to manmade disturbances.
Cultural landscapes are fragile and need to be maintained by man; especially in dry climates, abandonment
PART TWO / DISCOVERY AND SPOLIATION OF THE BIOSPHERE
management of patterns
type of management
FIGURE 4 Gullied hills in central Italy—a typical example
of processes of erosion acting on a densely cultivated area.
These processes, acting as refuge areas, ensure a high degree
of biodiversity.
Currently, neither species management on a site-by-site basis nor ecosystem management using single species are considered to be effective. Landscape ecology does not consider
single sites in isolation but rather the interactions between
the various components. One of the most useful ways of
studying ecosystems is to use a calibrated scale (Fig. 5).
Indeed, a landscape can be considered as a geographically defined box containing not only spontaneous patterns
and processes but also man-made ones. Watersheds can be
considered as landscapes, but it is also possible to identify
other functional units, such as ecotopes and micro-, meso-,
macro-, and megachores. For example, it seems useful to
study landscapes by applying the scale of watersheds, which
can be considered as multifunctional units in which flows of
water and the transfer of nutrients are distinctive processes.
Ecosystem management can also make effective use of
umbrella species—that is, those that condition the functioning of the entire system. Thus, in the prairies of North
America, the bison was a key species in that its grazing,
trampling, and excrement modified the whole system. A
similar role is performed by many ungulates in the savannas of Africa.
When a key species disappears, the whole system collapses and changes. Thus, it was probably the disappearance
of the bison that caused the disappearance of many grasses
and their subsequent replacement by shrubs.
management of species
management of processes
management of functional areas
FIGURE 5 Ecosystem management can take place on at least
four organizational levels. Each level requires instruments that
are attuned to a particular spatial and temporal context.
443
213-A1678 8/6/01 10:07 AM Page 444
ALMO FARINA
Principles for the Conservation
of Nature Reserves
Various principles deriving both from the theory of island biogeography (MacArthur and Wilson, 1967) and from
the combination of paradigms in landscape ecology allow
us to outline procedures for the conservation of areas of interest. First, it is important to estimate the minimum surface
area involved. A tropical forest, for example, must be large
enough to allow biodiversity. In temperate environments,
even forests of a modest size support a high degree of biodiversity. Contiguous areas conserve more inner-forest species than disjointed ones, even when the total surface area
is the same. In forest environments, patches that are separate but nevertheless near to each other have a greater number of species than patches that are far apart; also, patches
that are disjointed but linked by corridors of protected areas are more conducive to biodiversity than patches that
are completely isolated.
Granted similar environmental values, circular protected areas are preferable to elongated ones since the latter have a higher ratio between the perimeter and the surface area. These principles draw attention to the importance
of surface area, the shapes of the patches, and the connection and the development of the ecotones.
Disturbance Regimes and Conservation
Increasingly, knowledge of the patterns of land mosaics
appears to be an important element in the conservation of
nature. In particular, large-scale disturbance regimes play a
fundamental role in maintaining ecological processes (Baker, 1992). It is clear that disturbance regimes need to take
place within large areas so as to affect only a part of the
system. Indeed, in the case of fires, it is important that
many patches should survive so as to create a shifting mosaic and enable disturbed patches to recover via secondary
succession. Climax communities can thus be conserved if
they are surrounded by a buffer of younger successional
communities.
It is nevertheless unrealistic to hope that large areas will
remain undisturbed by human activity, and thus the role of
landscape ecology is to study a whole range of variables in
order to evaluate the ability of different mosaics—forest
remnants, land under cultivation, river systems, and urban
areas and infrastructures—to support a reasonable degree
of biodiversity. One of the main rules in delimiting a protected area is to ensure that it will include a zone of a certain minimum size in which disturbance dynamics can operate. In other words, the area should contain groups of
patches that are subjected to a disturbance regime and that
will ensure that the system is relatively stable in the short
term. This should result in the creation of a shifting mosaic
that has the greatest chance of guaranteeing ecological complexity. The boundaries of a nature reserve should coincide
with the outer limits of disturbances, such as the line beyond
444
which floods do not extend or at which fires burn themselves out naturally. It is advisable to avoid identifying a nature reserve with just one type of habitat; furthermore, it is
desirable, where possible, to impose certain restrictions on
activities beyond the boundaries of the reserve (Halladay
and Gilmour, 1995) so as to enable the various processes
and species to perform their functions fully.
Corridors and Conservation
Corridors, or narrow strips of a given environment that
are wedged in between different environments, are very
popular among planners but give rise to some perplexity
among ecologists (Harrison, 1992). As often happens in a
complex system, the function of a corridor does not always
coincide with patterns such as hedgerows, rows of trees, and
watercourses. In some cases, these structures are genuine
corridors, but in others they do not function as corridors
even though they exhibit patterns associated with them.
The more mobile species tend to move along preferential routes, but it is not always possible to identify these
by applying human-scale criteria. Morphological structures
such as canyons and mountain ridges often attract animals,
for example, because they are easy to travel along. Hares
tend to move along tracks made by domesticated animals,
and the same is true of wolves and many other small mammals. Certain large carnivores, such as pumas, do not expect to find much prey in corridors, and therefore they move
along them relatively quickly. For many animals, knowing
their territory is essential for survival. This is true both of
herbivores that move from one grazing site to another and
of predators that systematically hunt for their prey in their
territory.
The Conservation of Habitat
Fragments and Relict Populations
In environments that are dominated by man, it is very
common to find fragmented zones inhabited by more or less
isolated populations whose survival depends on the degree
of connectivity between the patches (Fig. 6). At the local
level, populations can die out in certain patches; however,
provided that this does not happen at the same time in all
of them, the species is guaranteed to survive at the mosaic
level. The size, shape, and quality of patches have a major
influence on reproduction and the risk of predation. The
qualitative character of a patch is not just a function of its
shape; indeed, two patches of similar size but with different
shapes have notably diverse ratios between surface area and
volume.
Ecological Conservation and
the Conservation of Processes
When using the paradigms of landscape ecology, it is easier and more realistic to pursue conservation policies directed at flows and processes rather than to concentrate on
VOLUME IV / THE LIVING WORLD
213-A1678 8/6/01 10:07 AM Page 445
LANDSCAPES AND THEIR ECOLOGICAL COMPONENTS
a
b
a landscape. Once the widely held misconceptions concerning species conservation are cast aside, the conservation of
both biological and ecological diversity becomes a strategic
objective. Granted that this does in fact occur, modern economic models can coexist with biodiversity in such a way as
to guarantee solutions that are ecologically valid. In other
words, in a world that is dominated by man, it is essential
to find disturbance regimes that mimic the natural disturbances that trigger environmental dynamics and that also
promote a satisfactory level of economic activity. Indeed,
the socioeconomic crises that have affected various societies at different periods in history and in different parts of
the earth have generally been provoked by the sudden desynchronization of ecological and socioeconomic processes.
◗
................................................
Methodological Approaches
to Studying Landscapes
c
FIGURE 6 Various strategies for increasing the degree of
connectivity between isolated areas. (a) Creation of a corridor,
which permits the survival of many species in disjointed areas.
(b) Improving the quality of the environment in intermediary
areas. (c) Expanding the areas by creating buffer zones.
the conservation of a given species, at least when there are
no particularly important species to conserve. It often happens that a species is present for short periods ranging from
a few days to a few years; in this case, it will often be associated with particular stages of ecological succession. Furthermore, the rarity of a species is often a function of its
demographic cycles, and it is possible for a species to be
rare on the margins of its habitat and very common in the
center. For example, the conservation of the migratory flows
of birds from the western Palearctic cannot involve protecting just one single species but must involve identifying the
areas in which the birds stop during migration. Most land
migratory birds stop in environments other than those in
which they winter and reproduce, and, as with warblers and
thrushes, like open areas such as grassland or shrub-rich
savanna, both of which offer ample opportunities for feeding. On the contrary, some birds are specialized during reproduction but become generalists during migration.
Hierarchical Patterns within Landscapes
and the Conservation of Biodiversity
When a landscape is viewed as a hierarchical system, its
various components exhibit a range of dynamics that vary
according to the closeness of the relationship. It is therefore impossible to conserve these systems without taking
this consideration into account, and in doing so one tends
to conserve all elements of the biodiversity contained within
PART TWO / DISCOVERY AND SPOLIATION OF THE BIOSPHERE
The spatial processing of data is a vital aspect of landscape
ecology and requires a series of instruments, some highly
sophisticated, that are able to collect and process geographical and ecological information. The instruments and techniques range from remote sensing to geographic information systems (GISs).
Analysis is carried out on at least two main levels. The
first is concerned with the shape and size of the patches that
make up a landscape, whereas the second is concerned with
the characteristics of the land mosaic (Fig. 7). Consequently,
it is important to consider the ratio between the length of
the perimeter and the surface area of patches as well as their
shape and size. An analysis of a land mosaic must first involve measuring the diversity between patches as well as
their nearness and the extent of their aggregation. The distances separating the patches are particularly important
with regard to estimating the degree of connectivity that any
one system is able to guarantee for a given organism. Furthermore, greater distance necessarily implies an increase
1
C
2
3
4
a
C1
C
B1
B
C
5
b
C2
C3
B
B2
c
FIGURE 7 A quantitative analysis of a landscape can involve
analyzing the single patches (a), analyzing the mosaic as a
whole (b) and studying the various types of vegetation cover
(in the example the A patches represent land under cultivation
and the B patches represent woods), or a more sophisticated
approach (c) in which the structural characteristics of each
patch are studied, together with the various types of cover
(C1–C3, B1, and B2).
445
213-A1678 8/6/01 10:07 AM Page 446
ALMO FARINA
in the risk of predation and a reduction in the likelihood of
vector transport (the transport of matter or energy through
the agency of physical forces of the movement of animals).
Some techniques are based on an analysis of electronic images and are a function of the characteristics of single pixels. Such analyses can be carried out using data that are expressed in raster format (in which a process is represented
by cells, pixels, or discrete spatial units).
The concept of lacunarity in a fractal distribution (Mandelbrot, 1975) provides the basis for an interesting approach
to studying spatial patterns. An object with low lacunarity
does not change when it is moved, whereas one whose gap
distribution has a high degree of heterogeneity takes on a
different spatial configuration when it is moved (Farina,
1998). It therefore seems useful to study environmental heterogeneity in terms of lacunarity. Since the analysis of edges
is particularly important in this context, the complexity of
a landscape can be studied not only with reference to the
heterogeneity of the patches but also with reference to that
of the edges (Fig. 8).
shape/size
irregularity of the form
Fractal Patterns in Landscapes
The spatial distribution of organisms and their aggregation reflects the heterogeneity of landscapes. Indeed, the
distribution of organisms is determined by the unevenness
of the distribution of resources and, therefore, of biomasses. Fractal geometry (Mandelbrot, 1975) offers a new
and promising way of measuring this complexity.
In environments that have been modified by man, regular geometric or linear shapes predominate. In a natural
system, however, the irregularity of the shapes precludes an
Euclidean approach, which is unsuitable for describing environmental complexity. Fractal geometry is considered to
have many promising applications, including the analysis
not only of patterns but also of phenomena such as the unevenness of the distribution of organisms, their movements,
and, more generally, their behavior.
The term fractal was coined by Mandelbrot in 1975 to refer to an irregular object whose irregularities are exhibited
on every scale. A fractal can be taken to mean an object or a
pattern with a highly convoluted character. When a fractal
object exhibits broad-scale patterns that repeat on smaller
scales, such an object is said to demonstrate self-similarity.
There are two different types of fractals: regular ones and
random ones. With the former, the self-similarity is the same
on every level; in other words, the object is a copy of itself
on every scale. The latter category, on the other hand, includes natural fractals such as clouds, coastlines, and organism abundance. However, self-similarity that is identifiable only statistically is a scale-dependent concept. For
example, coasts are fractal objects, and their length depends
on the scale of resolution that is used. Fractal geometry has
a broad range of applications that include such areas of the
natural sciences as geology, hydraulics, soil texture, dynam-
446
spatial relationships
diversity
FIGURE 8 Information that is quantifiable at the patch level
refers to shape and size, the irregularity of the shapes, the spatial relationships between the single patches, and the diversity
between the types of cover.
ics, microbe transport, and plant structure. Fractal analytical techniques are particularly useful for studying objects
which, due to their scalar properties, exhibit special complications. Thus, the abundance of an organism is a function of the scale with which environmental information is
aggregated—that is, the degree of resolution that is used.
Geographical Information Systems
A GIS uses technology with which it is possible to process spatial data with computers and data input instruments
such as digitizers and scanners and then express these data
in graph form with printers and plotters. A good GIS generally makes use of digital cartography, remote sensing, and
database management (Burrough, 1986).
GISs are considered indispensable for analyzing phenomena such as vegetation patterns, the distribution and
movement of animals in a landscape, changes in the use of
VOLUME IV / THE LIVING WORLD
213-A1678 8/6/01 10:07 AM Page 447
LANDSCAPES AND THEIR ECOLOGICAL COMPONENTS
soil, and the modeling of ecological processes (Tomlin, 1990;
Coulson et al., 1991; Maguire et al., 1991). There are two
types of GISs: vectorial ones and raster ones. The former
give a more accurate image of objects that reflects their
natural contours, whereas the latter breaks the object into
discrete parts or cells (pixels). The advantage of the raster
system is mainly the ease with which information can be
handled. Indeed, groups of cells that form part of a matrix
can be aggregated, compared with their neighbors, and then
reaggregated for different degrees of resolution. Furthermore, information from satellite images is in raster format
and is relatively easy to handle with a raster GIS. The disadvantage of this system, however, is that it uses a great
deal of memory space to store information.
Remote Sensing
The patterns in a landscape are generally easy to identify on satellite images, especially when 10 10-m sensors
are used. The degree of resolution that is available for analyzing large geographical areas varies from 10 to 30 m per
pixel and is sufficient to provide information on the patterns of the landscape under observation. It is indeed possible to gather a great deal of information from large areas,
and remote sensing techniques can highlight transition
zones between various environments. The complexity of
the edges is a good indicator of the processes taking place
in a given area. Remote sensing techniques make it possible
to locate the position on the ground of patterns identified
by sensors mounted on satellites. These global positioning
systems (GPSs), which were developed for military purposes and for air and sea navigation, are now available at an
affordable price (Leick, 1990; Hofmann-Wellenhof et al.,
1993) and offer great potential for application in the field of
ecology (Trimble Navigation, 1994). Using triangulation, a
group of satellites are able to identify the rover’s whereabouts on the ground with a degree of accuracy ranging
from 25 m to a few centimeters, depending on the configuration of the hardware. Thus, it is possible, due to GPSs, to
designate areas, locate animals, and track their movements.
Similarly, it is possible to scan an environment and take a
census of birds by locating the sightings (Farina, 1998). The
information is stored in real –time, and then at the end of
the work session it is transferred to a GIS for processing.
This technique makes it possible to carry out ground-level
observation of patterns such as watercourses, coastlines, or
ecotones.
References Cited
ALLEN, T. F. H., and HOEKSTRA, T. W. (1992). Toward a Unified
Ecology. Columbia Univ. Press, New York.
ALLEN, T. F. H., and STARR, T. B. (1982). Hierarchy. Perspectives
for Ecological Complexity. Univ. of Chicago Press, Chicago.
BAKER, W. L. (1992). The landscape ecology of large disturbances
PART TWO / DISCOVERY AND SPOLIATION OF THE BIOSPHERE
in the design and management of nature reserves. Landscape
Ecol. 7, 181–194.
BURROUGH, P. A. (1986). Principles of Geographic Information
Systems for Land Resources Assessment. Clarendon/Oxford
Univ. Press, Oxford/New York.
CLEMENTS, F. C. (1897). Peculiar zonal formations of the great
plains. Am. Nat. 31, 968.
CLEMENTS, F. C. (1905). Research Methods in Ecology. University
Publishing, Lincoln, NE.
COULSON, R. N., LOVELADY, C. N., FLAMM, R. O., SPRADLING,
S. L., and SAUNDERS, M. C. (1991). In Quantitative Methods in
Landscape Ecology (M. G. Turner and R. H. Gardner, Eds.),
pp. 153 –172. Springer-Verlag, New York.
DAUBENMIRE, R. (1968). Plant Communities. A Textbook of Plant
Synecology. Harper & Row, New York.
DELCOURT, H. R., and DELCOURT, P. A. (1988). Quaternary landscape ecology: Relevant scales in space and time. Landscape
Ecol. 2, 23 – 44.
DI CASTRI, F., HANSEN, A. J., and HOLLAND, M. M. (1988). A
new look at ecotones: Emerging international projects on landscape boundaries. Biol. Int. 17 (Special issue), 1–163.
FARINA, A. (1998). Principles and Methods in Landscape Ecology.
Chapman & Hall, London.
FARINA, A., TOMASELLI, M., and FERRARI, C. (1986). Studi sui
rapporti tra vegetazione e Microtus nivalis (Martins) nell’Appennino reggiano. Prime osservazioni (Mammalia, Rodentia).
Boll. Zool. 53, 86.
FORMAN, R. T. T. (1995). Land Mosaics. The Ecology of Landscapes and Regions. Cambridge Univ. Press, Cambridge, UK.
FORMAN, R. T. T., and GODRON, M. (1986). Landscape Ecology.
Wiley, New York.
GARDNER, R. H., MILNE, B. T., TURNER, M. G., and O’NEILL,
R. V. (1987). Neutral models for the analysis of a broad-scale
landscape patterns. Landscape Ecol. 1, 19 –28.
GILPIN, M., and HANSKI, I. (Eds.) (1991). Metapopulation Dynamics: Empirical and Theoretical Investigations. Academic Press
for the Linnean Society of London, London.
GREEN, B. H., SIMMONS, E. A., and WOLTJER, I. (1996). Landscape conservation. Some steps towards developing a new
conservation dimension. A draft report of the IUCN–CESP
landscape Conservation Working Group. Department of Agriculture, Horticulture, and Environment, Wye College, Ashford, Kent, UK.
GRIGGS, R. F. (1914). Observations on the behaviour of some species on the edges of their ranges. Bull. Torrey Bot. Club 41,
25 – 49.
HALLADAY, D., and GILMOUR, D. A. (Eds.) (1995). Conservation
Biodiversity outside Protected Areas. The Role of Traditional
Agro-ecosystems. IUCN, Cambridge, UK.
HANSEN, A. J., and DI CASTRI, F. (Eds.) (1992). Landscape
Boundaries. Consequences for Biotic Diversity and Ecological
Flows. Springer-Verlag, New York.
HANSEN, A. J., DI CASTRI, F., and NAIMAN, R. J. (1992). Ecotones:
What and Why? In Landscape Boundaries. Consequences for
Biotic Diversity and Ecological Flows (A. J. Hansen and F. Di
Castri, Eds.), pp. 9 – 46. Springer-Verlag, New York.
HANSKI, I., and GILPIN, M. (1991). Metapopulation dynamics:
Brief history and conceptual domain. Biol. J. Linnean Soc. 42,
3 –16.
447
213-A1678 8/6/01 10:07 AM Page 448
ALMO FARINA
HARRISON, R. L. (1992). Toward a theory of inter-refuge corridor
design. Conserv. Biol. 6, 293 –295.
HOFMANN-WELLENHOF, B., LICHTENEGER, H., and COLLINS, J.
(1993). Global Positioning System, Theory and Practice, 2nd
ed. Springer-Verlag, New York.
HOLLAND, M. M. (1988). SCOPE /MAB technical consultations
on landscape boundaries. Report of a SCOPE /MAP workshop ecotones. In A New Look at Ecotones: Emerging International Projects on Landscape Boundaries (F. Di Castri, A. J.
Hansen, and M. M. Holland, Eds.). International Union of Biological Sciences, Paris.
HOLLAND, M. M., RISSER, P. G., and NAIMAN, R. J. (1991). Ecotone. The Role of Landscape Boundaries in the Management
and Restoration of Changing Environments. Chapman & Hall,
London.
HUMBOLDT, VON A. (1807). Indeen zu Einer Geographie der Pflanzen nebst einem Gemälde der Tropenlander. Tubingen.
KOLASA, J., and PICKETT, S. T. A. (Eds.) (1991). Ecological Heterogenity. Springer-Verlag, New York.
LEICK, A. (1990).GPS Satellite Surveying. Wiley, New York.
LEVINS, R. (1970). Extinction. In Some Mathematical Questions in
Biology. Lectures in Mathematics in the Life Sciences (M. Gertenshaubert, Ed.), pp. 77–107. American Mathematical Society, Providence, RI.
LIVINGSTON, B. E. (1903). The distribution of the upland societies
of Kent County. Michigan Bot. Gazette 35, 36 –55.
MACARTHUR, R. H. (1984). Geographical Ecology, Patterns in the
Distribution of Species. Princeton Univ. Press, Princeton, NJ.
MACARTHUR, R. H., and WILSON, E. O. (1967). The Theory of Island Biogeography. Princeton Univ. Press, Princeton, NJ.
MAGUIRE, D. J., GOODCHILD, M. F., and RHIND, D. W. (Eds.)
(1991). Geographical Information Systems. Longman, Harlow,
UK.
MANDELBROT, B. B. (1975). Les Objects Fractals: Forme, Hasard
et Dimension. Flammarion, Paris.
MERRIAM, G. (1984). Connectivity: A fundamental characteristic
of landscape pattern. In Methodology in Landscape Ecological
Research and Planning: Proceedings of the First International
Seminar of the International Association of Landscape Ecology
(IALE), pp. 5 –15. Roskilde, Denmark.
NAIMAN, R. J., and DECAMPS, H. (1990). The Ecology and Management of Aquatic–Terrestrial Ecotones. Parthenon, Park
Ridge, IL.
NAVEH, Z. (1987). Biocybernetic and thermodynamic perspectives of landscape functions and land use patterns. Landscape
Ecol. 1, 75 – 83.
448
NAVEH, Z., and LIEBERMAN, A. S. (1994). Landscape Ecology.
Theory and Application, 2nd ed. Springer-Verlag, New York.
ODUM, E. P. (1959). Fundamentals of Ecology, 2nd ed. Saunders,
Philadelphia.
O’NEILL, R. V., DEANGELIS, D. L., WAIDE, J. B., and ALLEN,
T. F. H. (1986). A Hierarchical Concept of Ecosystems. Princeton Univ. Press, Princeton, NJ.
PICKETT, S. T. A., and WHITE, P. S. (1985). The Ecology of Natural
Disturbance and Patch Dynamics. Academic Press, Orlando,
FL.
PLACKTER, H., and ROSSLER, M. (1995). Cultural landscapes: Reconnecting culture and nature. In Cultural Landscapes of Universal Value (B. van Droste, P. Plackter and H. Rossler, Eds.).
Fischer in cooperation with UNESCO, New York.
PULLIAM, H. R. (1988). Source–sink, and population regulation.
Am. Nat. 132, 652 – 661.
PULLIAM, H. R. (1996). Sources and sinks: Empirical evidence and
population consequences. In Population Dynamics in Ecological Space and Time (O. E. Rhodes, R. K. Chesser, and M. H.
Smith, Eds.), pp. 45 – 69. Univ. of Chicago Press, Chicago.
RICKLEFS, R. E. (1973). Ecology. Chiron Press, Newton, MA.
RISSER, P. G. (Ed.) (1995). Understanding and managing ecotones. Ecol. Int. 22.
STAUFFER, D. (1985). Introduction of Percolation Theory. Taylor
& Francis, London.
TOMLIN, C. D. (1990). Geographic Information Systems and Cartographic Modelling. Prentice Hall, Englewood Cliffs, NJ.
Trimble Navigation (1994). GPS Mapping Systems, General Reference. Trimble Navigation, Sunnyvale, CA.
TURNER, M. G. (1987). Spatial simulation of landscape changes in
Georgia: A comparison of 3 transition models. Landscape
Ecol. 1, 29 –36.
TURNER, M. G., GARDNER, R. H., and O’NEILL, R. V. (1995). Ecological dynamics at broad scales. Bioscience 29 (Suppl.).
WATKINSON, A. R., and SUTHERLAND, W. J. (1995). Source, sinks
and pseudo-sinks. J. Anim. Ecol. 64, 126 –130.
WEAVER, J. E., and CLEMENTS, F. C. (1929). Plant Ecology. McGraw–Hill, New York.
WIENS, J. A., and MILNE, B. T. (1989). Scaling of landscape in landscape ecology, or landscape ecology from a beetle’s perspective.
Landscape Ecol. 3, 87–96.
ZIFF, R. (1986). Test scaling exponents for percolation-cluster
perimeters and ecology. Phys. Rev. Lett. 56, 545 –548.
ZONNEVELD, I. S. (1995). Land Ecology: An Introduction to Landscape Ecology as a Base for Land Evaluation, Land Management and Conservation. SPB Academic, Amsterdam.
VOLUME IV / THE LIVING WORLD